Effect of calcium propionate and monensin on in vitro digestibility and gas production

An evaluation of the effects of monensin and calcium propionate on the in vitro kinetics of gas production, digestibility, carbon dioxide, and minor gas production on different days was performed using the ruminal fluid from eight Suffolk lamb donors, after receiving additives for 1, 10, and 20 days. Treatments consisted of a control ration of 40% grain; 30 mg/kg of monensin in a diet with 40% grain; 10 g/kg calcium propionate in a diet with 30% grain; and the combination of both additives in a diet with 30% grain. The gas production was measured up to 72 h of incubation and all incubation procedures were repeated three times on days 1, 10, and 20. On incubation day 20, the volume and production of methane and minor gases were measured. There was an interaction between calcium propionate and monensin for maximum gas production, in vitro dry matter digestibility (IVDMD), carbon dioxide, and minor gases. Monensin reduced gas production on days 1 and 20, whereas calcium propionate increased gas production (Vm) on day 1. The rate of gas production (s) was reduced by calcium propionate on day 1 and by the combination of additives on day 10. Lag time was reduced by monensin on day 10; however, it declined linearly with the feeding time of the additives. Monensin had no effect on IVDMD (62.29 vs. 62.24%), while calcium propionate increased the IVDMD (60.00 vs. 64.53%). The inclusion of monensin increased CO2; however, the combination of monensin and calcium propionate had no effect on CO2 production. Monensin reduced methane (25.37 vs. 20.29%) and increased CO2. None of the additives showed consistent effects on the kinetic parameters of in vitro gas production over time. The treatments with monensin and calcium propionate showed a significant reduction in methane production, with a higher fermentation efficiency since the IVDMD was increased. Both additives are a strategy to consider to reduce methane emissions without affecting the ruminal fermentation.


Introduction
To maximize the production efficiency and reduce metabolic problems, it is a common practice to include feed additives in rations for high-producing animals. Certain products could have more advantages, such as reducing the greenhouse gas production without affecting animal performance. Ionophores are an alternative to reduce ruminal methane production (Ellis et al., 2012); monensin is a predominant antibiotic given to ruminants (Kim et al., 2014a), although its use is restricted in many countries. Alternatively, the use of calcium propionate may change the acetate to propionate ratio (Ferraro et al., 2009), allowing for a reduction in the grain level of the ration (Lee et al., 2012). The combination of both additives could reduce methane emissions (Moss et al., 2000;Wang et al., 2009) without affecting ruminant productivity and could be an interesting feeding option.
It is well known that monensin increases the production of propionate (Ellis et al., 2012); therefore, its combination with a calcium propionate could be a better alternative for reducing methane emissions. It is important to have a previous evaluation of the ruminal additives before their use in production that can be performed with an in vitro gas production technique, which is a precise and fast procedure (Getachew et al., 2004). Therefore, the objective of this study was to evaluate monensin, with or without calcium propionate, in complete diets in which grain was reduced to 100 g/kg using in vitro gas techniques to estimate CO 2 and minor gas production (CH 4 , N 2 , H 2 and others), and incubating at different days of the additive adaptation of the inoculum donors.

Material and Methods
The present investigation was performed in Texcoco, Estado de Mexico,Mexico (19°29'23"N and 98°53'37"W), following the guidelines of the local Academic Committee, under the Law of Animal Production of the State of Mexico.
Eight male Suffolk sheep (24 kg body weight, six months old) were ruminally cannulated and divided into four treatments; then, they were fed the experimental diets with or without additives (Table 1) and used as donors of rumen fluid. The animals were housed and fed in individual pens. Before the beginning of the experiment, the animals were fed a 50:50 (silage:concentrate) ration. Then, they were adapted to the experimental rations gradually during a 10-day period. The experiment lasted twenty days after the adaptation period. Sampling was performed three times during the experimental period on days 1, 10, and 20, in which the animals were fed diets with additives.
The sheep were fed 1.2 kg of an experimental diet twice daily (at 08.00 and 15.00 h) and water was provided ad libitum. Experimental diets were offered over 20 days and ruminal fluid was collected on days 1, 10, and 20.
The dietary treatments consisted of a control ration with 40% grain; 30 mg/kg of monensin sodium (Rumensin 200 g Elanco Animal Health, Mexico D.F.) in a diet with 40% grain; 10 g/kg calcium propionate (65%, Paniplus Mexico) in a diet with 30% grain; and 30 mg/kg of monensin sodium plus 10 g/kg calcium propionate in a diet with 30% grain. Diets were analyzed for dry matter, organic matter, crude protein, and ether extract according to AOAC (1990) and neutral detergent fiber procedures (Van Soest et al., 1991) (Table 1).
Rumen fluid was collected (100 mL) using a Tygon tube connected to a vacuum pump and a 1000-mL Erlenmeyer flask. Rumen fluid was transported to the laboratory in sterile plastic containers (150 mL) at 39 °C and the rumen fluid obtained from both donors for each treatment was obtained individually and then mixed for later use.
Ninety millilitres of reduced mineral solution and rumen fluid (9:1) were added to amber flasks (120 mL) containing 0.5 g of each diet and incubated under anaerobic conditions in a water bath at 39 °C. Each incubation was performed in triplicate, using each experimental diets as a substrate. Gas production was measured up to 72 h of incubation using the in vitro gas production procedure described by Menke and Steingas (1988).
After incubation (72 h), the residual was filtered to determine in vitro dry matter digestibility. All incubation procedures were repeated three times on days 1, 10, and 20. Gas volume data and incubation time were used to obtain the parameters of the kinetics of gas production: maximum volume of gas produced (Vm), lag phase (L), and rate of gas production (S), using the logistic model Vo = Vm / (1 + exp (2-4 * s * (t L) ) described by Pitt et al. (1999).
On incubation day 20, the volume and production of CO 2 after 24 h of fermentation were measured volumetrically, estimating the methane and minor gases by the difference in these values (Singh and Mohini, 1999). Thirty millilitres of reduced mineral solution and rumen fluid (2:1) were added into amber flasks (60 mL) containing 0.25 g of the each diet. Gas production was measured at 0, 6, 12, and 24 h with a 150-mL glass syringe to determine the production of CO 2 and minor gases. Flasks (60 mL) were prepared with 40 mL 1 N potassium hydroxide to capture the gas and then titrated with 0.55 N hydrochloric acid using 1 mL potassium hydroxide, one drop of barium chloride, and one drop of rezarsurine to determine CO 2 mmol/g = (B -S) × N × 1000/2 (weight sample), in which B = mL HCl used in the blank, S = mL HCl used in the  Table 1 -Experimental diets and chemical compositions sample (mL), and N = normality of HCl. The percentage of CO 2 was estimated as a function of the total maximum volume.
Results were analyzed as a completely randomized design with a 2 × 2 factorial arrangement (factors were the additive source at two levels) with the repeated measure procedure (days 0, 10, and 20). A Tukey's test was used to compare treatment means using the JMP software (Sall et al., 2012).
The model included the fixed effects of treatments and its interaction. Data were analyzed with the repeated measures over time (Littell et al., 1998) according to the model: Yijklm is the dependent variable; μ is the overall mean; α i is the fixed effect of calcium propionate (i =1, 2); τ k is the fixed effect of monensin level (k = 1, 2); (ατ) ik is the calcium propionate × monensin; d k (ij) is the random effect of the k-th subject (animal) nested within treatment (i.e., the calcium propionate × monensin level interaction); γm is the fixed effect of the day of evaluation (m = 1, 2, 3); (αγ)im is the Ca-Pr × day of evaluation interaction; (τγ)km is the monensin level × day of evaluation interaction; (ατγ)ikm is the Ca-Pr × monensin × day of evaluation interaction; and εijklm is the random error.

Results
There was an interaction (P<0.05) between calcium propionate and monensin for maximum gas production (Vm), in vitro dry matter digestibility (IVDMD), CO 2 , and methane and minor gases (Table 2). Monensin reduced Vm on days 1 and 20 (P<0.05), whereas calcium propionate increased Vm on day 1 (P<0.05). The rate of gas production was reduced by calcium propionate on day 1 by 1 and by the combination of additives on day 10 (P<0.05). Lag time was reduced by monensin on day 10 (P<0.01) and declined linearly with adaptation time (Figure 1).
The inclusion of monensin increased CO 2 (74.62 vs. 79.70%), whereas the combination of monensin and calcium propionate avoided this increment (interaction P<0.008). Methane production was reduced with monensin (25.37 vs. 20.29%), whereas its combination with calcium propionate had no effect (Table 2). Likewise, calcium propionate did not affect methane production. A longer duration of additive feeding improved the digestibility, Ca-Pr -calcium propionate; SEM -standard error of the mean; Vm -maximum volume of gas produced; s -rate of gas production; L -lag phase; IVDMD -in vitro dry matter digestibility.
Means with different letters are statistically different (P<0.05).
Table 2 -Effect of monensin and calcium propionate on in vitro gas production kinetics, digestibility, methane, and carbon dioxide production rate of gas production, and lag time parameters of the lambs (Figure 3 and Table 3).

Discussion
Because of the antibiotic nature of monensin, we expected it to have an inhibitory effect on all kinetic parameters of gas production and digestibility. The effect of monensin is a function of the dosage. Two in vitro experiments reported that monensin decreased gas production and that the doses were higher than those used in animal feeds (Smith, 2010;Kim et al., 2014a). In contrast, Singh and Mohani (1999) did not find any effect on in vitro gas production using 50 or 100 mg/d of monensin in rumen fluid donors. The lack of effects of monensin in this study agrees with molecular studies that did not reveal large shifts in the relative abundance of Gram-positive bacteria in response to monensin from in vivo samples (Kim et al., 2014b).
The absence of effects on in vitro dry matter agrees with findings of Singh and Mohani (1999), without changes in in vivo organic and dry matter digestibility. Smith (2010) reported that the in vitro digestibility was not affected by monensin, even when microbial growth was depressed. Similar results were reported by Garcia et al. Schelling (1984) concluded that monensin have no effect on cellulose digestibility when animals were adapted for 21 days, which agrees with the current study, because sheep had more time receiving the ionophore. Consequently, the linear response on time confirms the adaptation for all additives on day 20.
It is not clear why the addition of calcium propionate increased in vitro gas production; its inclusion probably produced a change in pH and osmotic pressure. Greater amylolytic activity has been observed in similar conditions (Mendoza et al., 1999), because the dissociation of calcium propionate contributes to the production of volatile fatty acids. Lee et al. (2012) reported that the addition of calcium propionate in diets for lambs resulted in a higher proportion of propionic acid in lambs. Miranda et al. (2016) did not find the stimulatory effect of calcium propionate on the volume of gas production with similar grain levels. Other glycogenic substrates, such as glycerol, reduced in vitro gas production (Young et al., 2011;Ferraro et al., 2009); however, the kinetics of degradation was affected by the adaptation period. Young et al. (2011) observed that after   an adaptation, ruminal bacteria can easily ferment glycerol, whereas ruminal fluid from unadapted donors shows a considerably longer lag phase. The estimation indicates that monensin reduced methane production by 10% in comparison with the control, which is similar to that reported in other studies (Guan et al., 2006;Johnson et al., 2009;Ellis et al., 2012), and can be explained by the effect on bacteria susceptible to ionophores, which are the main producers of acetate, butyrate, and H 2 , and CO 2 substrates for methanogenesis (Moss et al., 2000). Singh and Mohani (1999) reported reductions of methane between 23 and 28% with slightly higher doses than those reported in this experiment. In contrast, Kim et al. (2014a) observed a dramatic reduction in methane (53%) with doses not biologically useful for animals.
The addition of calcium propionate could reduce methane production because it is metabolized into propionic acid. The formation of propionate requires hydrogen and does not produce CO 2 (D' Mello, 2000); however, this effect was not observed by Miranda et al. (2016), who studied grain level and calcium propionate. Callaway and Martin (1996) incubated corn with organic acid plus monensin (5 mg/kg), which increased CO 2 production because of lactate fermentation to propionate via the succinate propionate pathway, with carbon dioxide as the end-product. It has been demonstrated that ionophores select for major propionate producers such as Selenomonas rumiantium, which ferment lactate using that pathway (Mackie et al., 1984).
There is evidence that monensin reduces Gram-positive bacteria groups, rumen protozoa, and fungi, whereas some bacteria strains are more sensitive than others and some are resistant, mainly due to the structure of the cell wall of organisms (Ellis et al., 2012, Mendoza et al., 1993. Research indicates that monensin may decrease the number of protozoa by 4 to 63% (Shelling, 1984).

Conclusions
The use of monensin and calcium propionate during a given period of time shows an improved in vitro dry matter digestibility, reducing methane production, resulting in a better fermentative efficiency. Both additives are a viable nutritional strategy for the reduction of Greenhouse gases such as methane.